Q & A: New Novell CEO Ron Hovsepian And Channel Chief Steve Erdman

I believe Messman was doing a good job. The Linux part was growing 20% per year. Unfortunately, the rest was sluggish. What could Novell expect of a product nearing end of life? Novell did well in the old days on the server/network side of things which Microsoft ignored. Now, many customers find no need to use Novell's stuff if they use Windows or Linux because Novell provided nothing essential. Novell is switching to the Linux game but they cannot do it fast enough because their customers are locked in. They can only convert a customer to Linux when the customer is ready, tired of waiting for Vista, starting a new setup, or sick of Windows. They found it difficult to switch to Linux in their own internal operations.

I am not a selling genius, but it seems to me they have to gently wean customers off the old stuff and find new customers. They can do things like giving systems away to schools to build mind-share in the population, pump up their Linux community, advertise, put on some showy events, and do these sooner rather than later. The 3D desktop would spark interest if they put it on TV and showed it to more of the media. They can convert folks who are tired of waiting for the eye-candy from Microsoft. Schools and medium to large businesses would probably be more interested if Novell did more with the hardware using thin clients. So far, Novell are doing a lot of java stuff which promotes software development and integration but wastes a fair amount of CPU power. Any customers looking to save money probably are not well served that way. The thin client, especially with multi-seat clients really multiplies the benefit of hardware expenditures. RedHat helps thin client users in a big way with K12LTSP and Ubuntu does it for EdUbuntu. Will Novell do the same?

For an example, consider a system I recently designed for a school wanting to upgrade from Windows 98 and a 10 mb/s LAN. Two four way Opteron servers with 6 gB RAM each can serve 150 simultaneous users easily. Users sit six on each AMD64 thin client on a gigabit LAN. There are only two machines needing software installation and maintenance. There are 27 machines needing occasional dusting. Cost of hardware is less than $200 per seat and the software cost is a few hours of work to install and the usual user and software maintenance on two servers. Linux thick clients would cost several times as much because of the extra CPUs, boxes, hard drives, cabling, switches, and a huge increase in software maintenance. How many more customers could Novell gather if they could offer state of the art software and hardware so cheaply to a larger market? While such customers would expect to pay much less than the $100/seat which is common, it would also cost Novell much less to serve these customers through their partners. It is a win for everyone.